This project aims to design personalized, data-driven policy recommendations for education programs, for example, math course-taking plans in high school. We leverage recent advances in personalized medicine, known as optimal (dynamic) treatment regimes, to recommend the best treatment option for each individual in a way that maximizes a desirable educational outcome. In addition to optimizing utility, we incorporate critical considerations such as feasibility, interpretability, and fairness into the recommendation models.

This work is supported by NSF’s EHR Core Research Building Capacity in STEM Education Research (ECR: BCSER) program.

Publications/Working Papers

  • Pan, C., Li, Y., & Suk, Y. (2026). Learning feasible optimal treatment regimes for personalized decision-making. PsyArXiv. [Preprint]
  • Suk, Y., Park, C., Pan, C., & Kim, K. (2024). Fair and robust estimation of heterogeneous treatment effects for optimal policies in multilevel studies. PsyArXiv. [Preprint] [R Code]
  • Suk, Y., & Park, C. (2023). Designing optimal, data-driven policies from multisite randomized trials. Psychometrika, 88. 1171-1196. [Journal Article] [Preprint] [R Code]

Recent Conferences/Seminars

  • Li, Y., Pan, C., & Suk, Y. (2026, April). Designing feasible optimal treatment regimes for personalized education. The American Educational Research Association (AERA), Los Angeles, CA, U.S.
  • Pan, C., Li, Y., & Suk, Y. (2026, April). Enhancing the interpretability of heterogeneous treatment effects using Kolmogorov Arnold Network. The American Educational Research Association (AERA), Los Angeles, CA, U.S.
  • Li, Y., Pan, C., & Suk, Y. (2025, Oct). Improving the feasibility of optimal treatment regimes for personalized education. The 3rd Annual Conference of Advanced Quantitative Methods and Analytics for Public Policy Support (AQMAPPS), New York, NY, U.S.
  • Pan, C., Li, Y., & Suk, Y. (2025, Oct). Investigating interpretable optimal treatment regimes using Kolmogorov-Arnold Networks. The 3rd Annual Conference of Advanced Quantitative Methods and Analytics for Public Policy Support (AQMAPPS), New York, NY, U.S.
  • Pan, C., Li, Y., & Suk, Y. (2025, Oct). Investigating interpretable optimal treatment regimes using Kolmogorov-Arnold Networks. The Society for Research on Educational Effectiveness (SREE), Chicago, IL, U.S.
  • Li, Y., Pan, C., & Suk, Y. (2025, Oct). Improving the feasibility of optimal treatment regimes for personalized education. The Society for Research on Educational Effectiveness (SREE), Chicago, IL, U.S.
  • Suk, Y., Park, C., Pan., C., & Kim, K. (2025, Oct). Fair and robust estimation of heterogeneous treatment effects in multilevel studies. The Society for Research on Educational Effectiveness (SREE), Chicago, IL, U.S.
  • Suk, Y., Park, C., Pan., C., & Kim, K. (2025, July). Fair and robust estimation of heterogeneous treatment effects in multilevel studies. The International Conference on Education Research (ICER), Seoul, South Korea.
  • Pan, C. & Suk, Y. (2025, May). Designing realistic and interpretable optimal treatment regimes for personalized education. The American Causal Inference Conference (ACIC), Detroit, MI, U.S. -Pan, C., Li, Y., & Suk, Y. (2025, May). Learning interpretable optimal treatment regimes Using Kolmogorov-Arnold Networks. The Psychology@TC Student Research Conference, NY, U.S.
  • Li, Y., Pan, C., & Suk, Y. (2025, May). Improving the feasibility of optimal treatment regimes for personalized education. The Psychology@TC Student Research Conference, NY, U.S.
  • Pan, C., Li, Y., & Suk, Y. (2025, April). Learning interpretable optimal treatment regimes Using Kolmogorov-Arnold Networks. The Foundations of Data Science Workshop, Columbia University, Data Science Institute, NY, U.S.
  • Pan, C., & Suk, Y. (2025, Apr.). Designing optimal dynamic treatment regimes using TMLE for personalized math course-taking plans. The American Educational Research Association (AERA), Denver, CO, U.S.
  • Suk, Y., & Pan, C. (2024, Sep.). Designing personalized math course-taking plans in high school using optimal treatment regimes. The Society for Research on Educational Effectiveness (SREE), Baltimore, MD, U.S.
  • Suk, Y., Kim, K., & Park, C. (2024, Apr.). Towards fair and personalized education policy: Reducing racial and state disparities in advanced math courses. The American Educational Research Association (AERA), Philadelphia, PA, U.S.
  • Suk, Y., & Park, C. (2023, May). Designing optimal, data-driven educational policies from multisite randomized trials. The American Causal Inference Conference (ACIC), Austin, TX, U.S.